Operating system application programming interfaces and methods of using operating systems

Operating system functions are defined as objects that are collections of data and methods. The objects represent operating system resources. The resource objects can be instantiated and used across process and machine boundaries. Each object has an associated handle that is stored in its private state. When an application requests a resource, it is given a second handle or pseudo handle that corresponds with the handle in the object's private state. The second handle is valid across process and machine boundaries and all access to the object takes place through the second handle. This greatly facilitates remote computing. In preferred embodiments, the objects are COM objects and remote computing is facilitated through the use of Distributed COM (DCOM) techniques. Other embodiments of the invention provide legacy and versioning support by identifying each resource, rather than the overall operating system, with a unique identifier that can specified by an application. Different versions of the same resource have different identifiers. This ensures that applications that need a specific version of a resource can receive that version. This also ensures that an application can specifically request a particular version of a resource by using its unique identifier, and be assured of receiving that resource. Other embodiments of the invention provide legacy support by intercepting calls for operating system functions and transforming those calls into object calls that can be understood by the resource objects. This is accomplished in preferred embodiments by injecting a level of indirection between an application and an operating system.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
RELATED APPLICATION

[0001] This is a continuation application of, and priority is claimed to U.S. Patent Application No. 06/334,079, filed on Jun. 16, 1999, the disclosure of which is incorporated by reference herein.

TECHNICAL FIELD

[0002] This invention relates to operating systems and to methods of program use to interact with operating systems.

BACKGROUND OF THE INVENTION

[0003] Operating systems typically include large numbers of callable functions that are structured to support operation on a single host machine. When an application executes on the single host machine, it interacts with the operating system by making one or more calls to the operating system's functions.

[0004] Although this method of communicating with an operating system has been adequate, it has certain shortcomings. One such shortcoming relates to the increasing use of distributed computing, in which different computers are programmed to work in concert on a particular task. Specifically, operating system function libraries can severely limit the ability to perform distributed computing.

[0005] FIG. 1 illustrates the use of functions in prior art operating systems. FIG. 1 shows a system 20 that includes an operating system 22 and an application 24 executing in conjunction with the operating system 22. In operation, the application 24 makes calls directly into the operating system when, for example, it wants to create or use an operating system resource. As an example, if an application wants to create a file, it might call a “CreateFile” function at 26 to create the file. Responsive to this call, the operating system returns a “handle” 28. A “handle” is an arbitrary identifier, coined by the operating system to identify a resource that is controlled by the operating system. In this example, the application uses handle 28 to identify the newly created file resource any time it makes subsequent calls to the operating system to manipulate the file resource. For example, if the application wants to read the file associated with handle 28, it uses the handle when it makes a “ReadFile” call, e.g. “ReadFile (handle)”. Similarly, if the application wants to write to the file resource it uses handle 28 when it makes a “WriteFile” call, e.g. “WriteFile (handle)”.

[0006] One problem associated with using a handle as specified above is that the particular handle that is returned to an application by the operating system is only valid for the process in which it is being used. That is, without additional processing the handle has no meaning outside of its current process, e.g. in another process on a common or different machine. Hence, the handle cannot be used across process or machine boundaries. This makes computing in a distributed computing system impossible because, by definition, distributed computing takes place across process and machine boundaries. Thus, current operating systems lack the ability to name and manipulate operating system resources on remote machines.

[0007] Another problem with traditional operating system function libraries is that individual functions cannot generally be modified without jeopardizing the operation of older versions of applications that might depend on the particular characteristics of the individual functions. Thus, when an operating system is upgraded it typically maintains all of the older functions so that older applications can still use the operating system.

[0008] In prior art operating systems, a function library essentially defines a protocol for communicating with an operating system. When operating systems are upgraded, the list of functions that it provides typically changes. Specifically, functions can be added, removed, or changed. This changes the protocol that is used between an application and an operating system. As soon as the protocol is changed, the chances that an application will attempt to use a protocol that is not understood by the operating system, and vice versa increase.

[0009] Prior art operating systems attempt to deal with new versions of operating systems by using so-called version numbers. Version numbers are assigned to each operating system. Applications can make specific calls to the operating system to ascertain the version number of the operating system that is presently in use. For example, when queried by an application, Windows NT 4 returns a “4” and Windows NT 5 returns a “5”. The application must then know what specific protocol to use when communicating with the operating system. In addition, in order for an operating system to know what operating system version the application was designed for, a value is included in the application's binary. The operating system can then attempt to accommodate the application's protocol.

[0010] The version number system has a couple of problems that can adversely affect functionality. Specifically, a typical operating system may have thousands of functions that can be called by an application. For example, Win32, a Microsoft operating system application programming interface, has some 8000 functions. The version number that is assigned to an operating system then, by definition, represents all of the possibly thousands of functions that an operating system supports. This level of description is undesirable because it does not provide an adequate degree of resolution. Additionally, some operating systems can return the same version number. Thus, if the operating systems are different (which they usually are), then returning the same version number can lead to operating errors. What is needed is the ability to identify different versions of operating systems at a level that is lower than the operating system level itself. Ideally, this level should be at or near the function level so that a change in just one or a few functions does not trigger a new version number for the entire operating system.

[0011] The present invention arose out of concerns associated with providing improved flexibility to operating systems. Specifically, the invention arose out of concerns associated with providing operating systems that are configured for use in distributed computing environments, and that can easily support legacy applications and versioning.

SUMMARY OF THE INVENTION

[0012] Operating system functions are defined as objects that are collections of data and methods. The objects represent operating system resources. The resource objects can be instantiated and used across process and machine boundaries. Each object has an associated handle that is stored in its private state. When an application requests a resource, it is given a second handle or pseudo handle that corresponds with the handle in the object's private state. The second handle is valid across process and machine boundaries and all access to the object takes place through the second handle. This greatly facilitates remote computing. In preferred embodiments, the objects are COM objects and remote computing is facilitated through the use of Distributed COM (DCOM) techniques.

[0013] Other embodiments of the invention provide legacy and versioning support by identifying each resource, rather than the overall operating system, with a unique identifier that can specified by an application. Different versions of the same resource have different identifiers. This ensures that applications that need a specific version of a resource can receive that version. This also ensures that an application can specifically request a particular version of a resource by using its unique identifier, and be assured of receiving that resource.

[0014] Other embodiments of the invention provide legacy support by intercepting calls for operating system functions and transforming those calls into object calls that can be understood by the resource objects. This is accomplished in preferred embodiments by injecting a level of indirection between an unmodified application and an operating system.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG. 1 is a diagram that illustrates a prior art operating system.

[0016] FIG. 2 is a diagram of a computer that can be used to implement various embodiments of the invention.

[0017] FIG. 3 is a diagram of one exemplary operating system architecture.

[0018] FIG. 4 is a high level diagram of an operating system having a plurality of its resources defined as objects and distributed across process and machine boundaries.

[0019] FIG. 5 is a diagram of an exemplary architecture in accordance with one embodiment of the invention.

[0020] FIG. 6 is a diagram that illustrates operational aspects of one embodiment of the invention.

[0021] FIG. 7 is a diagram of one exemplary operating system architecture.

[0022] FIG. 8 is a diagram of one exemplary operating system architecture.

[0023] FIG. 9 is a diagram of one exemplary operating system architecture.

[0024] FIG. 10 is a flow diagram that describes processing in accordance with one embodiment of the invention.

[0025] FIG. 11 is a block diagram that illustrates one aspect of an interface factoring scheme.

[0026] FIGS. 12-15 are diagrams of interface hierarchies in accordance with one embodiment of the invention.

DETAILED DESCRIPTION

[0027] Overview

[0028] Various examples will be given in the context of Microsoft's Win32 operating system application programming interface and function library, commonly referred to as the “Win32 API.” Although this is a specific example, it is not intended to limit the principles of the invention to only the Win32 function library or, for that matter, to Microsoft's operating systems. The Win32 operating system is described in detail in a text entitled Windows 95 WIN32 Programming API Bible, authored by Richard Simon, and available through Waite Group Press.

[0029] In accordance with one embodiment of the invention, one or more of an operating system's resources are defined as objects that can be identified and manipulated by an application through the use of object-oriented techniques. Generally, a resource is something that might have been represented in the prior art as a particular handle “type.” Examples of resources include files, windows, menus and the like.

[0030] Preferably, all of the operating system's resources are defined in this way. Doing so provides flexibility for distributed computing and legacy support as will become apparent below. By defining the operating system resources as objects, without reference to process-specific “handles,” the objects can be instantiated anywhere in a distributed system. This permits responsibility for different resources to be split up across process and machine boundaries. Additionally, the objects that define the various operating system resources can be identified in such a way that applications have no trouble calling the appropriate objects when they are running. This applies to whether the applications know they are running in connection with operating system resource objects or not. If applications are unaware that they are running in connection with operating system resource objects, e.g. legacy applications, a mechanism is provided for translating calls for the functions into object calls that are understood by the operating system resources objects.

[0031] In addition, factorization schemes are provided that enable an operating system's functions to be re-organized and redefined into a plurality of object interfaces that have methods corresponding to the functions. In preferred embodiments, the interfaces are organized to leverage advantages of interface aggregation and inheritance.

[0032] Preliminarily, FIG. 2 shows a general example of a desktop computer 130 that can be used in accordance with the invention. Various numbers of computers such as that shown can be used in the context of a distributed computing environment. In this document, computers are also referred to as “machines”.

[0033] Computer 130 includes one or more processors or processing units 132, a system memory 134, and a bus 136 that couples various system components including the system memory 134 to processors 132. The bus 136 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. The system memory 134 includes read only memory (ROM) 138 and random access memory (RAM) 140. A basic input/output system (BIOS) 142, containing the basic routines that help to transfer information between elements within computer 130, such as during start-up, is stored in ROM 138.

[0034] Computer 130 further includes a hard disk drive 144 for reading from and writing to a hard disk (not shown), a magnetic disk drive 146 for reading from and writing to a removable magnetic disk 148, and an optical disk drive 150 for reading from or writing to a removable optical disk 152 such as a CD ROM or other optical media. The hard disk drive 144, magnetic disk drive 146, and optical disk drive 150 are connected to the bus 136 by an SCSI interface 154 or some other appropriate interface. The drives and their associated computer-readable media provide nonvolatile storage of computer-readable instructions, data structures, program modules and other data for computer 130. Although the exemplary environment described herein employs a hard disk, a removable magnetic disk 148 and a removable optical disk 152, it should be appreciated by those skilled in the art that other types of computer-readable media which can store data that is accessible by a computer, such as magnetic cassettes, flash memory cards, digital video disks, random access memories (RAMs), read only memories (ROMs), and the like, may also be used in the exemplary operating environment.

[0035] A number of program modules may be stored on the hard disk 144, magnetic disk 148, optical disk 152, ROM 138, or RAM 140, including an operating system 158, one or more application programs 160, other program modules 162, and program data 164. A user may enter commands and information into computer 130 through input devices such as a keyboard 166 and a pointing device 168. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner, or the like. These and other input devices are connected to the processing unit 132 through an interface 170 that is coupled to the bus 136. A monitor 172 or other type of display device is also connected to the bus 136 via an interface, such as a video adapter 174. In addition to the monitor, personal computers typically include other peripheral output devices (not shown) such as speakers and printers.

[0036] Computer 130 commonly operates in a networked environment using logical connections to one or more remote computers, such as a remote computer 176. The remote computer 176 may be another personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to computer 130, although only a memory storage device 178 has been illustrated in FIG. 2. The logical connections depicted in FIG. 2 include a local area network (LAN) 180 and a wide area network (WAN) 182. Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets, and the Internet.

[0037] When used in a LAN networking environment, computer 130 is connected to the local network 180 through a network interface or adapter 184. When used in a WAN networking environment, computer 130 typically includes a modem 186 or other means for establishing communications over the wide area network 182, such as the Internet. The modem 186, which may be internal or external, is connected to the bus 136 via a serial port interface 156. In a networked environment, program modules depicted relative to the personal computer 130, or portions thereof, may be stored in the remote memory storage device. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used.

[0038] Generally, the data processors of computer 130 are programmed by means of instructions stored at different times in the various computer-readable storage media of the computer. Programs and operating systems are typically distributed, for example, on floppy disks or CD-ROMs. From there, they are installed or loaded into the secondary memory of a computer. At execution, they are loaded at least partially into the computer's primary electronic memory. The invention described herein includes these and other various types of computer-readable storage media when such media contain instructions or programs for implementing the steps described below in conjunction with a microprocessor or other data processor. The invention also includes the computer itself when programmed according to the methods and techniques described below.

[0039] For purposes of illustration, programs and other executable program components such as the operating system are illustrated herein as discrete blocks, although it is recognized that such programs and components reside at various times in different storage components of the computer, and are executed by the data processor(s) of the computer.

[0040] General Operating System Object Architecture

[0041] FIG. 3 shows an exemplary group of objects generally at 30 that represent a plurality of operating system resources 32, 34, 36, 38 within operating system 22. Resource 32 is a file resource, resource 34 is a window resource, resource 36 is a font resource, and resource 38 is a menu resource. The objects contain methods and data that can be used to manipulate the object. For example, file object 32 might include the methods “CreateFile”, “WriteFile”, and “ReadFile”. Similarly, window object 34 might include the methods “CreateWindow”, “CloseWindow” and “FlashWindow”. Any number of objects can be provided and are really only limited by the number of functions that exist in an operating system, and/or the way in which the functions are factored as will become apparent below. In various embodiments, it has been found advantageous to split the functions into a plurality of objects based upon a logical relationship between the functions. One advantage of doing this is that it facilitates computing in a distributed system and limits the complexity of doing so. Specifically, by dividing the functions logically between various objects, only objects having the desired functionality are instantiated on a remote machine. For example, if all of the functions that are associated with displaying a window on a display device are contained within a single object, then only that object need be instantiated on a remote display machine, e.g. a handheld device. Although it is possible for all of the functions of an operating system to be represented by a single object, this would add to overhead during remote processing. The illustrated architecture is particularly useful for applications that are “aware” they are operating in connection with resource objects. These applications can make specific object calls to the resource objects without the need to intercept and translate their calls, as will be discussed below.

[0042] Although any suitable object model can be used to define the operating system resources, it has been found particularly advantageous to define them as COM objects. COM objects are well known Microsoft computing mechanisms and are described in a book entitled Inside OLE, Second Edition 1995, which is authored by Kraig Brockschmidt. In COM, each object has one or more interfaces that are represented by the plug notation used in FIG. 3. An interface is a group of semantically related functions or methods. All access to an object occurs through member functions of an interface. Representing the operating system resources as objects provides an opportunity to redefine the architecture of current operating systems, and to provide new architectures that improve upon the old ones.

[0043] One advantage of representing resources as COM objects comes in the remote computing environment. Specifically, when COM objects are instantiated throughout a distributed computing system, Distributed COM (DCOM) techniques can be used for remote communication. DCOM is a known communication protocol developed by Microsoft.

[0044] FIG. 4 shows an exemplary distribution of an operating system's resources across one process boundary and one machine boundary in a distributed computing system. In the described example, resource object 48 is instantiated in-process (i.e. inside the application's process), resource object 50 is instantiated in another process on the same machine (i.e. local), and resource object 52 is instantiated on another machine (i.e. remote). The instantiated resource objects are used by the application 24 and constitute a translation layer between the application and the operating system. Specifically, the application makes object calls on the resource objects. The resource objects, in turn, pass the calls down into the operating system in a manner that is understood by the operating system. One way of doing this is through the use of handle/pseudo handle pairs discussed in more detail below.

[0045] In order to use the resource objects, the application must first be able to communicate with them. In one embodiment where the operating system resources comprise COM objects, communication takes place through the use of known DCOM techniques. Specifically, in the local case where resource 50 is instantiated across a process boundary, DCOM provides for an instantiated proxy/stub pair 54 to marshal data across the process boundary. The remote case also uses a proxy/stub pair 54 to marshal data across the process and machine boundaries. In addition, an optional proxy manager 56 can be instantiated or otherwise provided to oversee communication performed by the proxy/stub pair, and to take measures to reduce unnecessary communication. Specifically, one common proxy manager task is to cache remote data to avoid unnecessary communication. For example, in the Win32 operating system, information can be cached to improve the re-drawing of remote windows. When a BeginPaint( ) call is made, it signals the beginning of a re-draw operation by creating a new drawing context resource. In order to be available remotely, this resource has to be wrapped by an object. Rather than creating a new object instance on each re-draw operation, an object instance can be cached in the proxy manager and re-used for the re-draw wrapper

[0046] Translation Layer

[0047] FIG. 5 shows a translation layer 58 comprising resource objects 32, 34, 36, and 38. Translation layer 58 is interposed between an application 24 that is configured to make resource object calls, and an operating system 22 that is not configured to receive the resource object calls. In this example, application 24 is not a legacy application because those applications directly call functions in the operating system. Translation layer 58 works in concert with application 24 so that the application's resource object calls can be used by the object to call functions of the operating system.

[0048] FIG. 6 shows one way that translation layer 58 translates resource object calls from the application 24 into calls to operating system functions. Here, the operating system resources are defined as COM objects that have one or more interfaces that are called by the application. Because the COM objects can be instantiated either in process, locally, or remotely, the standard handle that was discussed in the “Background” section cannot be used. Recall that the reason for this is that the handle is only valid in its own process, and not in other processes on the same or different machines. To address and overcome the limitations that are inherent with the use of this first handle, aspects of the invention create a second or “pseudo” handle and associate it with the first handle. The second handle is preferably valid universally, outside the process of the first handle. This means that the second handle is valid across multiple machine and process boundaries. The application uses the second handle instead of the first handle whenever it creates or manipulates an operating system resource.

[0049] In operation, an application initially calls a resource object in the translation layer 58 when it wants to create a resource to use. An application may, for example, call “CreateFile” on a file object to create a file. The application is then passed a pseudo-handle 60 instead of the first handle 28 for the file object. The first handle 28 is stored in the object instance's private state, i.e. it remains with its associated object. This means that the file object has its own real handle 28 that it maintains, and the application has a pseudo handle 60 that corresponds to the real handle. Application 24 makes object calls to the object of interest using the pseudo-handle 60. The object takes the pseudo-handle, retrieves the corresponding handle 28 and uses it to call functions in the operating system. The application uses the pseudo-handle 60 for all access to the operating system resource. In a preferred embodiment, pseudo-handle 60 is an interface pointer that points to an interface of the object of interest.

[0050] With an appropriate pseudo-handle, an application is free to access any of the resources that are associated with an object that corresponds to that handle. This means that the application uses the pseudo-handle 60 to make subsequent calls to, in this example, the file object. For example, calls to “ReadFile” and “WriteFile” now take place using the pseudo handle 60. When the application makes a call using the pseudo handle 60, the object determines the real handle that corresponds to the pseudo-handle. Any suitable method can be used such as a mapping process. If the object is in process, then the call gets passed down to the operating system 22 using first handle 28 as shown. If the object is local or on another machine, then communication takes place with the object at its current location across process and machine boundaries. Where the operating system resources are defined as COM objects, DCOM techniques can be used to call across process and machine boundaries.

[0051] Legacy Application Support

[0052] FIGS. 7 and 8 show two different architectures that can be used in connection with legacy applications. FIG. 7 includes an operating system that is the same as the one described in connection with FIG. 5. FIG. 8 includes an operating system that is the same as the one described in connection with FIG. 3.

[0053] Recall that legacy applications are those that call operating system functions instead of objects. These types of applications do not have any way of knowing that they are running in connection with a system whose resources are defined as objects. Hence, when an application calls a function, it “believes” that the function is supported by and accessible through the operating system. The syntax of the function calls, however, is not understood by the objects. Embodiments of the invention translate the syntax of the function calls into syntax that is understood by the objects. In accordance with one embodiment, application calls are intercepted and transformed before reaching the operating system. The transformed calls are then used to call the appropriate object using the syntax that it can understand. Then the object passes the calls into the operating system as was described above in connection with FIG. 6.

[0054] In one implementation, a detour 60 is provided that implements a detour function. Detour 60 is interposed between the application and the operating system. When an application calls a function, detour 60 intercepts the call and transforms it into an object call. In preferred embodiments, detour 60 enables communication across at least one and preferably more process and machine boundaries for remote computing. Where the objects are COM objects, communication takes place through DCOM techniques discussed above.

[0055] To understand how one embodiment of detour 60 works, the following example is given. Syntactically, detour 60 changes the syntax of an application's call to an operating system function into one that is understood by an object. For example, a prior art call might use the following syntax to call “ReadFile”: ReadFile (handle, buffer, size), where “handle” specifies a file resource that is to be read. There are many different resources that can be read using the ReadFile function, e.g. a file, a pipe, and a socket.

[0056] When a prior art operating system is called in this manner, the operating system typically looks for the code that is associated with reading the particular type of resource that is specified by the handle, and then reads the resource using the code. One way prior art operating systems can do this is to have one lengthy “IF” statement that specifies the code to be used for each different type of resource. Thus, if a new resource is to be added, the “IF” statement must be modified to provide for that type of resource.

[0057] Detour 60 greatly streamlines this process by translating the “ReadFile” call syntax into one that can be used by the specified resource. So in this example, the original “handle” actually specifies an object. The new syntax for the object call is represented as “handle→ReadFile (buffer, size)”. Here, “handle” is the object and “ReadFile” is an object function or method. In COM embodiments, the “ReadFile” method of the handle object is accessed through the object's vtable in a known manner. This configuration allows an object to contain only the code that is specifically necessary to operate upon it. It need not contain any code that is associated with other types of objects. This is advantageous because new objects can be created simply by providing the code that is uniquely associated with it, rather than by modifying a lengthy “IF” statement. Each object is self-contained and does not impact or affect any of the other objects. Nor does its creation affect the run time of any other objects. Only those applications that need a specific object will have it created for their use. Another advantage is the ease with which objects can be accessed. Specifically, applications can access the various objects through the use of pseudo-handles which are discussed above.

[0058] Detour 60 constitutes but one way of making a syntactic transformation from one format that cannot be used with resource objects to a format that can be used with resource objects. This supports legacy applications that do not “know” that they are running on top of a system whose resources are provided as objects. So, to the application it appears as if its calls are working just the same as they ever did.

[0059] Detour Implementation

[0060] When an application is built, it links against a set of dynamic linked libraries or (DLLs). The DLLs contain code that corresponds to the particular calls that an application makes. For example, the call “CreateFile” is contained in a DLL called “kernel32.dll”. At run time, the operating system loads “kernal32.dll” into the address space for the application. Detour 60 contains a detour call for each call that an application makes. So, in this example, detour 60 contains a call “Detour_CreateFile”. The goal of detour 60 is to call the “Detour_CreateFile” called every time the application calls “CreateFile”. This provides a level of indirection when the application makes a call to the operating system. The indirection enables certain decisions to be made by detour 60 that relate to whether a call is made locally or remotely.

[0061] As an example, consider the following. If an application desires to use a “WriteFile” call to write certain data to a particular file remotely, but also to write certain other data to a file locally, then a redirected “Detour_WriteFile” call can determine that there is a local operation that must take place, as well as a remote operation that must take place. The “Detour_WriteFile” call can then make the appropriate calls to ensure that the local operation does in fact take place, and the appropriate calls to ensure that the remote operations do in fact take place.

[0062] One way of injecting this level of indirection into the call is to manipulate the call's assembly code. Specifically, portions of the assembly code can be removed and replaced with code that implements the detour. So, using the “CreateFile” call as an example, the first few lines of code comprising the “CreateFile” call are removed and replaced with a “jump” instruction that calls “Detour_CreateFile”. For those operating systems that do not natively implement ii resource objects, a trampoline 62 (FIG. 9) is provided and receives the lines of code that are removed, along with another jump instruction that jumps back to the original “CreateFile” call. Now, when application 24 calls “CreateFile”, detour 60 automatically calls “Detour_CreateFile”. If there is local processing that must take place, the “Detour_CreateFile” can call trampoline 62 to invoke the original local “CreateFile” sub-routine. Otherwise, if there is remote processing that must take place, the detour 60 can take the appropriate steps to ensure that remote processing takes place. In this manner, the detour 60 wedges between the application and the operating system with a level of indirection. The indirection provides an opportunity to process either locally or remotely.

[0063] One of the primary advantages of detour 60 in the COM embodiments is the remoting capabilities provided by DCOM. That is, because the operating system's resources are now modeled as COM objects, DCOM can be used essentially for free to support communication with local or remote processes or machines.

[0064] Linking Against Detours

[0065] One way that detours can be implemented is to modify the dynamic link library (DLL) that an application links against. Specifically, rather than link against DLLs and their associated functions, an application links directly against detour functions, e.g. “detour32.dll” instead of “kernel32.dll”. Here, “detour32.dll” contains the same function names as “kernel32.dll”. However, rather than providing the kernel's functionality, “detour32.dll” contains object calls. Thus, an application makes a function call to a function name in the “detour32.dll” which, in turn, makes an object call.

[0066] With the “detour.dll”, all of the function calls are translated into COM calls. The trampoline 62 is loaded and is hardwired so that it knows where to jump to the appropriate places in the kernal32.dll.

[0067] Version Support

[0068] Another aspect of the invention provides support for different versions of a resource within an operating system. Recall that in the prior art, operating system versions are simply represented by a version number. The version number represents the entire collection of operating system functions. Thus, a modification to a handful of operating system functions might spawn a new operating system version and version number. Yet, many if not most of the original functions remain unchanged. Because of this, version numbers provide an undesirable degree of description. In addition, recall that previous operating systems maintain vast function libraries that include all of the functions that an application might need. Function calls cannot be deleted because legacy applications might need them. This results in a large, bulky architecture of collective functions that is not efficient.

[0069] While the functionality of these functions must be maintained to support legacy applications, various embodiments of the invention do so in a manner that is much less cumbersome and much more efficient. Specifically, embodiments of the invention create the necessary resources for legacy applications only when they are needed by an application. The resources are defined as objects that are collections of data and methods. Each object only contains the methods that pertain to it. No other resources are created or maintained if they are not specifically needed by an application. This is made possible, in the preferred embodiment, through the use of COM objects that encapsulate the functionality of the requested resources.

[0070] Accurate version support is provided by the way in which object interfaces are identified. Specifically, each object interface has its own unique identifier. Each different version of a resource is represented by a different interface identifier. An application can specifically request a unique identifier when it wants a particular version of a resource.

[0071] One way of implementing this in COM is as follows. As background, every interface in COM is defined by an interface identifier, or IID that is formed by a globally unique identifier or “GUID”. GUIDs are numbers that are generated by the operating system and are bound by the programmer or a development tool to the interfaces that they represent. By programming convention, no two incompatible interfaces can ever have the same IID. One of the rules in COM that accompanies the use of these GUIDs is that if an interface changes in any way whatsoever, so too must its associated IID change. Thus, IIDs and interfaces are inextricably bound together and provide a way to uniquely identify the interface with which it is associated over all other interfaces in its operating universe.

[0072] In the present invention, every operating system function is implemented as a method of some interface that has its own assigned unique identifier. In the preferred embodiment, the unique identifier comprises a GUID or IID. Other unique identifiers can, of course, be used. An application that uses a set of functions now specifies unique identifiers that are associated with the functions. This assures the application that it will receive the exact versions of the functions or methods that it needs to execute. In addition, in those circumstances where the resources are instantiated across a distributed system, the unique identifiers are specified across multiple process and machine boundaries. In a preferred embodiment, the applications store the appropriate unique identifiers, GUIDs, or IIDs in their data segment.

[0073] One of the benefits of using unique identifiers or IIDs is that each represents the syntax and the semantics of an interface. If the syntax or the semantics of an interface changes, the interface must be assigned a new identifier or IID. By representing the operating system resources as COM objects that support these interfaces, each with their own specific identifier or IID, applications can be assured of the desired call syntax and semantics when specific interfaces are requested. Specifically and with reference to the COM embodiments, an application that knows it is operating on an operating system that has its resources defined as COM objects can call QueryInterface on a particular object. By specifying the IID in the QueryInterface call, the application can determine whether that object implements a specific version of a specific interface.

[0074] In addition, embodiments of the invention can provide an operating system with the ability to determine, based on the specified unique identifier, whether it has the resource that is requested. If it does not, the operating system can ascertain the location of the particular resource and retrieve it so that the application can have the requested resource. The location from which the resource is retrieved can be across process and machine boundaries. As an example, consider the following. If an application asks for a specific version of a “ReadFile” interface, and the operating system does not support that version, the operating system may know where to go in order to download the code to implement the requested functionality. Software code for the specific requested interface may, for example, be located on a web site to which the operating system has access. The operating system can then simply access the web site, download the code, and provide the resource to the application.

[0075] Linking Against Unique Identifiers

[0076] When an application is linked, it typically links against a set of DLL names and entry points in a known manner. The DLLs contain code that corresponds to the particular calls that an application will need to make. So for example, if an application knows that it is going to need the call “CreateFile”, it will link against the DLL name that includes the code for that call, e.g. “kernel32.dll”. At run time, a loader for the operating system loads “kernel32.dll” into the address space for the application. Linking against DLLs in this manner does not support versioning because there is no way to specify a particular version of a resource.

[0077] To address this and other problems, one embodiment of the invention establishes a library that contains unique identifiers for one or more interfaces, e.g. GUIDs, and the method offsets that are associated with the unique identifiers. The method offsets correspond to the vtable entry for the unique identifier. An application is then linked against the unique identifiers. For example, when an application is compiled, it is linked against one or more “.lib” or library files. A linker is responsible for taking the “.lib” files that have been specified by the application and looking for the functions or methods that are needed by the application. When the linker finds the appropriate specific functions, it copies information out of the “.lib” file and into the binary image of the application. This information includes the name of the DLL containing the functions, and the name of the function. Linking by GUID and method offset can be accomplished by simply modifying the library or “.lib” files by replacing the DLL names and function names with the GUIDs and method offsets. This change does not affect the application, operating system, or compiler. For example, DLL names typically have the form “xxxxxx.dll”. The GUID identifier, on the other hand, is represented as a hexadecimal string that is specified by a set of brackets “{ }”. The linker and the loader can then be modified by simply specifying that they should look for the brackets, instead of looking for the “xxxxxx.dll” form. This results in loading only those specific interfaces (containing the appropriate methods) that are needed for an application instead of any DLLs. This supports versioning because an application can specifically name, by GUID, the specific interfaces that it needs to operate. Accordingly, only those interfaces that constitute the specific version that an application requests are loaded.

[0078] Factorization

[0079] Factorization involves looking at a set of functions and reorganizing the functions into defined interfaces based upon some definable logical relationship between the functions. In the described embodiment of the invention, the existing functions of an operating system are factored and assigned to different interfaces, so that the functions are now implemented as interface methods. The interfaces are associated with objects that represent underlying operating resources such as files, windows, etc. In this context, an “object” is a data structure that includes both data and associated methods. The objects are preferably COM objects that can be instantiated anywhere throughout a remote computing system. Factoring the function calls associated with an operating system's resources provides independent operating system resources and promotes clarity. It also promotes effective, efficient versioning, and clean remoting of the resources.

[0080] FIG. 10 shows a flow diagram at 200 that describes factorization steps in accordance with one embodiment of the invention. Step 202 factors function calls into first interface groups based upon a first criteria. An exemplary first criteria takes into account the particular items or underlying resources associated with the operation of a function, or the particular manner in which a function behaves. For example, some functions might be associated only with a window resource in that they create a window or allow a window to be manipulated in some way. These types of functions are placed into a first group that is associated with windows. An exemplary first interface group might be designated IWin32 Window.

[0081] Step 204 factors the first groups into individual sub-groups based upon a second criteria. An exemplary second criteria is based upon the nature of the operation of a function on the particular item or resource with which it is associated. For example, by nature, some functions create resources such as windows, while other functions do not create resources. Rather, these other functions have an effect on, or operate in some manner on a resource after it has been created. Accordingly, step 204 considers this operational nature and assigns the functions to different sub-groups based upon operational differences. In one embodiment, the groups are factored into sub-groups by considering the call parameters and return values that the functions use. This permits factorization to take place based upon each function's use of a handle. As an example, consider the following five functions:

[0082] HANDLE CreateWindow( . . . );

[0083] int DialogBoxParam( . . . ,HANDLE, . . . );

[0084] int FlashWindow(HANDLE, . . . );

[0085] HANDLE GetProp (HANDLE, . . . )

[0086] int GetWindowText(HANDLE, . . . );

[0087] A loaded operating system resource is exported to the application as an opaque value called a kernel handle. Functions that create kernel handles (i.e., resources) are moved to a “factory” interface, and functions that then query or manipulate these kernel handles are moved to a “handle” interface. Accordingly, step 206 assigns the sub-groups to different object interfaces. For example, those functions that create a window are assigned into an interface sub-group called IWin32WindowFactory, while those functions that do not create a window, but rather operate on it in some way are assigned into an interface sub-group called IWin32WindowHandle. Each interface represents a particular object's implementation of its collective functions. Objects can now be created or instantiated that include interfaces that contain one or more methods that correspond to the functions. Objects can be instantiated anywhere in a remote computing environment.

[0088] In a further extension of the factorization, consideration is given to functions that act upon a number of different resources. For example, Win32 has several calls that synchronize on a specified handle. The specified handle can represent a standard synchronization resource, such as a mutual exclusion lock, or less common synchronization resources such as processes or files. By simply factoring the functions as described above, this relationship would be missed. For example, the synchronization calls would be placed in a IWin32SyncHandle interface, while the process and file calls would be placed in IWin32ProcessHandle and IWin32FileHandle interfaces, respectively. In order to capture the relationship between these functions though, the process and file interfaces should also include the synchronization calls. Because the process and file handles can be thought of as logically extending the functionality of the synchronization handle, the concept of interface inheritance can be used to ensure that this takes place. Accordingly, both the IWin32ProcessHandle and IWin32FileHandle will thus inherit from the IWin32SyncHandle interface. This means that the IWin32ProcessHandle and IWin32FileHandle interfaces contain all the methods of the IWin32SyncHandle interface, in addition to their own methods.

[0089] To assist in further understanding of the factorization scheme, the following example is given by considering again the five functions listed above. FIG. 11 constitutes a small but exemplary subset of the 130+ window functions in the Win32 operating system. The “CreateWindow( )” function creates a window. The remaining functions execute a dialog box, flash the window's title bar, query various window properties, and return the current text in the window title bar. These functions all operate on windows in some way and are first factored into a windows group. Next, the functions are further factored depending on their use of kernel handles (denoted by “HANDLE” in the above functions). Since “CreateWindow( )” creates a handle or window, it is factored into a factory sub-group called IWin32WindowFactory. Since the other functions do not create a window, but only operate on or relative to one, they are placed in a handle sub-group called IWin32WindowHandle. In a third step, the IWin32WindowHandle sub-group is further factored into IWin32WindowState and IWin32Property interfaces. The State and Property interfaces are said to compose the IWin32WindowHandle interface. This composition is modeled through interface aggregation. The dialog calls are factored into their own interface since they are logical extensions of the windows. This is modeled through interface inheritance. Interface aggregation and inheritance are discussed in more detail in the Brockschmidt text above.

[0090] To further assist in understanding the factorization scheme, FIGS. 12-15 are provided, as well as the factorization list below. FIGS. 12-15 lists the interface hierarchy and factoring of a subset of more than one thousand functions of the Win32 operating system. The subset contains the necessary Win32 functions to support three operating system-intensive applications: Microsoft PhotoDraw, the Microsoft Developers' Network Corporate Benefits sample, and Microsoft Research's Octarine. The first is a commercial image manipulation package, the second is a widely distributed sample three-tiered, client-server application, and the third is a prototype COM-based integrated productivity application. All obsolete Windows 3.1 (16-bit) calls have been placed in IWin16 interfaces. In implementation, the top-level call prototypes will mirror their Win32 counterparts, with the appropriate parameters replaced by interface pointers. Note that these calls can wrap lower-level methods that implement different parameters. For example, the lower level methods could return descriptive HRESULTs directly and the Win32 return types as OUT parameters. Additionally, ANSI API calls can be implemented as wrappers of their UNICODE counterparts. The wrappers will simply perform argument translation and then invoke the counterpart.

[0091] The factorization list below lists the interface hierarchy. Inheritance relationships are clearly shown by the connecting lines, while aggregation is pictured by placing one interface block within another. This section also lists the call factorization. In the factorization list, “X:Y” denotes that X inherits from Y, and “Y←X” denotes that X is aggregated into Y. 1 Factorization List Generic Handles IWin32Handle CloseHandle Atoms IWin32Atom GlobalDeleteAtom GlobalGetAtomNameA IWin32AtomFactory GlobalAddAtomA Clipboard IWin32Clipboard ChangeClipboardChain CloseClipboard GetClipboardData GetClipboardFormatNameA GetClipboardFormatNameW GetClipboardOwner GetClipboardViewer GetOpenClipboardWindow IsClipboardFormatAvailable SetClipboardData IWin32ClipboardFactory RegisterClipboardFormatA RegisterClipboardFormatW Console IWin32Console : IWin32SyncHandle GetConsoleMode GetNumberOfConsoleInputEvents PeekConsoleInputA ReadConsoleA ReadConsoleInputA SetConsoleMode SetStdHandle WriteConsoleA IWin32ConsoleFactory AllocConsole GetStdHandle Drawing IWin16DeviceContextFont : IWin16DeviceContext EnumFontFamiliesA EnumFontsW GetCharWidthA GetTextExtentPointA GetTextExtentPointW IWin16MetaFile : IWin16DeviceContext CloseMetaFile CopyMetaFileA DeleteMetaFile EnumMetaFile GetMetaFileA GetMetaFileBitsEx GetWinMetaFileBits PlayMetaFile PlayMetaFileRecord IWin16MetaFileFactory GetEnhMetaFileA SetEnhMetaFileBits SetMetaFileBitsEx IWin32Bitmap:IWin32GDIObject CreatePatternBrush GetBitmapDimensionEx GetDIBits SetBitmapDimensionEx SetDIBits SetDIBitsToDevice IWin32BitmapFactory CreateBitmap CreateBitmapIndirect CreateCompatibleBitmap CreateDIBSection CreateDIBitmap CreateDiscardableBitmap IWin32BrushFactory CreateBrushIndirect CreateDIBPatternBrushPt CreateHatchBrush CreateSolidBrush IWin32Colorspace DeleteColorSpace IWin32ColorspaceFactory CreateColorSpaceA IWin32Cursor DestroyCursor SetCursor IWin32CursorFactory GetCursor IWin32CursorUtility ClipCursor GetCursorPos SetCursorPos ShowCursor IWin32DeviceContext← IWin32DeviceContextFont, IWin32DeviceContextCoords, IWin32Path, IWin32DeviceContextProperties, IWin32ScreenClip AngleArc Arc ArcTo BitBlt Chord CreateCompatibleDC DeleteDC DrawEdge DrawEscape DrawFocusRect DrawFrameControl DrawIcon DrawIconEx DrawStateA DrawTextA DrawTextW Ellipse EnumObjects ExtFloodFill ExtTextOutA ExtTextOutW FillRect FillRgn FloodFill FrameRect FrameRgn GdiFlush GetCurrentObject GetCurrentPositionEx GetPixel GrayStringA GrayStringW InvertRect InvertRgn LineDDA LineTo MaskBlt MoveToEx PaintRgn PatBlt Pie PlgBlt PolyBezier PolyBezierTo PolyDraw PolyPolygon PolyPolyline Polygon Polyline PolylineTo Rectangle ReleaseDC ResetDCA RestoreDC RoundRect SaveDC ScrollDC SetPixel SetPixelV StretchBlt StretchDIBits TabbedTextOutA TextOutA TextOutW WindowFromDC IWin32DeviceContextCoordinates DPtoLP LPtoDP IWin32DeviceContextFactory CreateDCA CreateDCW CreateICA CreateICW CreateMetaFileA CreateMetaFileW IWin32DeviceContextFont EnumFontFamiliesExA GetAspectRatioFilterEx GetCharABCWidthsA GetCharABCWidthsFloatA GetCharABCWidthsW GetCharWidth32A GetCharWidth32W GetCharWidthFloatA GetFontData GetGlyphOutlineA GetGlyphOutlineW GetKerningPairsA GetOutlineTextMetricsA GetTabbedTextExtentA GetTextAlign GetTextCharacterExtra GetTextCharsetInfo GetTextColor GetTextExtentExPointA GetTextExtentExPointW GetTextExtentPoint32A GetTextExtentPoint32W GetTextFaceA GetTextMetricsA GetTextMetricsW SetMapperFlags SetTextAlign SetTextCharacterExtra SetTextColor SetTextJustification IWin32DeviceContextProperties GetArcDirection GetBkColor GetBkMode GetBoundsRect GetBrushOrgEx GetColorAdjustment GetColorSpace GetDeviceCaps GetMapMode GetNearestColor GetPolyFillMode GetROP2 GetStretchBltMode GetViewportExtEx GetViewportOrgEx GetWindowExtEx GetWindowOrgEx OffsetViewportOrgEx OffsetWindowOrgEx PtVisible RectVisible ScaleViewportExtEx ScaleWindowExtEx SetArcDirection SetBkColor SetBkMode SetBoundsRect SetBrushOrgEx SetColorAdjustment SetColorSpace SetDIBColorTable SetICMMode SetMapMode SetMiterLimit SetPolyFillMode SetROP2 SetStretchBltMode SetViewportExtEx SetViewportOrgEx SetWindowExtEx SetWindowOrgEx UpdateColors IWin32EnhMetaFile: IWin32DeviceContext CloseEnhMetaFile CopyEnhMetaFileA CreateEnhMetaFileA CreateEnhMetaFileW DeleteEnhMetaFile EnumEnhMetaFile GdiComment GetEnhMetaFileBits GetEnhMetaFileDescriptionA GetEnhMetaFileDescriptionW GetEnhMetaFileHeader GetEnhMetaFilePaletteEntries PlayEnhMetaFile PlayEnhMetaFileRecord IWin32EnhMetaFileFactory SetWinMetaFileBits IWin32FontFactory CreateFontA CreateFontIndirectA CreateFontIndirectW CreateFontW IWin32GDIObject DeleteObject GetObjectA GetObjectType GetObjectW SelectObject UnrealizeObject IWin32GDIObjectFactory GetStockObject IWin32Icon CopyIcon DestroyIcon GetIconInfo IWin32IconFactory CreateIcon CreateIconFromResource CreateIconFromResourceEx CreateIconIndirect CreateMenu IWin32Palette : IWin32GDIObject AnimatePalette GetNearestPaletteIndex GetPaletteEntries ResizePalette SelectPalette SetPaletteEntries IWin32PaletteFactory CreateHalftonePalette CreatePalette IWin32PaletteSystem GetSystemPaletteEntries GetSystemPaletteUse RealizePalette IWin32Path AbortPath BeginPath CloseFigure EndPath FillPath FlattenPath GetMiterLimit GetPath PathToRegion StrokeAndFillPath StrokePath WidenPath IWin32PenFactory CreatePen CreatePenIndirect ExtCreatePen IWin32Print : IWin32DeviceContext AbortDoc EndDoc EndPage Escape ExtEscape SetAbortProc StartDocA StartDocW StartPage IWin32Rect CopyRect EqualRect InflateRect IntersectRect IsRectEmpty OffsetRect PtInRect SetRect SetRectEmpty SubtractRect UnionRect IWin32Region : IWin32GDIObject CombineRgn EqualRgn GetRegionData GetRgnBox OffsetRgn PtInRegion RectInRegion SetRectRgn IWin32RegionFactory CreateEllipticRgn CreateEllipticRgnIndirect CreatePolyPolygonRgn CreatePolygonRgn CreateRectRgn CreateRectRgnIndirect CreateRoundRectRgn ExtCreateRegion IWin32ScreenClip : IWin32DeviceContext ExcludeClipRect ExcludeUpdateRgn ExtSelectClipRgn GetClipBox GetClipRgn IntersectClipRect OffsetClipRgn SelectClipPath SelectClipRgn Environment IWin32EnvironmentUtility FreeEnvironmentStringsA FreeEnvironmentStringsW GetEnvironmentStrings GetEnvironmentStringsW GetEnvironmentVariableW SetEnvironmentVariableA SetEnvironmentVariableW File IWin16File : IWin16Handle _hread _hwrite _lclose _llseek _lopen _lwrite IWin16FileFactory OpenFile _lcreat _lread IWin32File : IWin32AsyncIOHandle FlushFileBuffers GetFileInformationByHandle GetFileSize GetFileTime GetFileType LockFile LockFileEx ReadFile ReadFileEx SetEndOfFile SetFilePointer SetFileTime UnlockFile WriteFile WriteFileEx IWin32FileFactory CreateFileA CreateFileW OpenFileMappingA IWin32FileMapping: IWin32ASyncIOHandle MapViewOfFile UnmapViewOfFile IWin32FileMappingFactory CreateFileMappingA IWin32FileSystem CopyFileA CopyFileEx CopyFileW CreateDirectoryA CreateDirectoryExA CreateDirectoryExW CreateDirectoryW DeleteFileA DeleteFileW GetDiskFreeSpaceA GetDiskFreeSpaceEx GetDriveTypeA GetDriveTypeW GetFileAttributesA GetFileAttributesW GetFileVersionInfoA GetFileVersionInfoSizeA GetLogicalDriveStringsA GetLogicalDrives GetVolumeInformationA GetVolumeInformationW MoveFileA MoveFileEx MoveFileW RemoveDirectoryA RemoveDirectoryW SetFileAttributesA SetFileAttributesW UnlockFileEx VerQueryValueA IWin32FileUtility AreFileApisANSI CompareFileTime DosDateTimeToFileTime FileTimeToDosDateTime FileTimeToLocalFileTime FileTimeToSystemTime GetFullPathNameA GetFullPathNameW GetShortPathNameA GetShortPathNameW GetTempFileNameA GetTempFileNameW GetTempPathA GetTempPathW LocalFileTimeToFileTime SearchPathA SystemTimeToFileTime IWin32FindFile : IWin32ASyncIOHandle FindClose FindCloseChangeNotification FindFirstFileEx FindNextChangeNotification FindNextFileA FindNextFileW IWin32FindFileFactory FindFirstChangeNotificationA FindFirstChangeNotificationW FindFirstFileA FindFirstFileW Interprocess Communication IWin32DDE DdeAccessData DdeDisconnect DdeFreeDataHandle DdeFreeStringHandle DdeUnaccessData IWin32DDEFactory DdeClientTransaction DdeConnect DdeCreateStringHandleA IWin32DDEUtility DdeGetLastError DdeInitializeA ReuseDDElParam UnpackDDElParam IWin32Pipe : IWin32AsyncIOHandle PeekNamedPipe IWin32PipeFactory CreatePipe Keyboard IWin32Keyboard GetAsyncKeyState GetKeyState GetKeyboardState MapVirtualKeyA SetKeyboardState VkKeyScanA keybd_event IWin32KeyboardLayout ActivateKeyboardLayout IWin32KeyboardLayoutFactory GetKeyboardLayout Memory IWin16GlobalMemory : IWin16Memory GlobalFlags GlobalFree GlobalLock GlobalReAlloc GlobalSize GlobalUnlock IWin16GlobalMemoryFactory GlobalAlloc GlobalHandle IWin32Heap : IWin32Memory HeapAlloc HeapCompact HeapDestroy HeapFree HeapReAlloc HeapSize HeapValidate HeapWalk IWin32HeapFactory GetProcessHeap HeapCreate IWin16LocalMemory : IWin16Memory LocalFree LocalLock LocalReAlloc LocalUnlock IWin32LocalMemoryFactory LocalAlloc IWin16Memory IsBadCodePtr IsBadReadPtr IsBadStringPtrA IsBadStringPtrW IsBadWritePtr IWin32Memory IsBadCodePtr IsBadReadPtr IsBadStringPtrA IsBadStringPtrW IsBadWritePtr IWin32VirtualMemory : IWin32Memory VirtualFree VirtualLock VirtualProtect VirtualQuery VirtualUnlock IWin32VirtualMemoryFactory VirtualAlloc Module IWin32Module : IWin32Handle DisableThreadLibraryCalls EnumResourceNamesA FindResourceA FreeLibrary GetModuleFileNameA GetModuleFileNameW GetProcAddress LoadBitmapA LoadBitmapW LoadCursorA LoadCursorW LoadIconA LoadIconW LoadImageA LoadMenuA LoadMenuIndirectA LoadStringA SizeofResource IWin32ModuleFactory GetModuleHandleA GetModuleHandleW LoadLibraryA LoadLibraryExA LoadLibraryW Multiple Window Position IWin32MWP BeginDeferWindowPos DeferWindowPos EndDeferWindowPos Ole IWin32Ole CoDisconnectObject CoLockObjectExternal CoRegisterClassObject CoRevokeClassObject IWin32OleFactory BindMoniker CoCreateInstance CoGetClassObject CoGetInstanceFromFile CreateDataAdviseHolder CreateDataCache CreateILockBytesOnHGlobal CreateOleAdviseHolder CreateStreamOnHGlobal OleCreate OleCreateDefaultHandler OleCreateFromData OleCreateFromFile OleCreateLink OleCreateLinkFromData OleCreateLinkToFile OleGetClipboard OleLoad IWin32OleMarshalUtility CoMarshalInterface CoReleaseMarshalData CoUnmarshalInterface IWin32OleMoniker CreateGenericComposite CreateItemMoniker CreatePointerMoniker CreateURLMoniker MkParseDisplayName MonikerCommonPrefixWith MonikerRelativePathTo IWin32OleMonikerFactory CreateBindCtx CreateFileMoniker GetRunningObjectTable IWin32OleStg OleConvertIStorageToOLESTREAM OleSave ReadClassStg ReleaseStgMedium WriteClassStg WriteFmtUserTypeStg IWin32OleStgFactory StgCreateDocfile StgCreateDocfileOnILockBytes StgIsStorageFile StgOpenStorage IWin32OleStream GetHGlobalFromStream OleConvertOLESTREAMToIStorage OleLoadFromStream OleSaveToStream ReadClassStm WriteClassStm IWin32OleUtility CLSIDFromProgID CLSIDFromString CoCreateGuid CoFileTimeNow CoFreeUnusedLibraries CoGetMalloc CoInitialize CoRegisterMessageFilter CoTaskMemAlloc CoTaskMemFree CoTaskMemRealloc CoUninitialize GetClassFile GetHGlobalFromILockBytes IIDFromString OleGetIconOfClass OleInitialize OleIsRunning OleRegEnumVerbs OleRegGetMiscStatus OleRegGetUserType OleSetClipboard OleUninitialize ProgIDFromCLSID PropVariantClear RegisterDragDrop RevokeDragDrop StringFromCLSID StringFromGUID2 StringFromIID OpenGL IWin32GL glBegin glClear glClearColor glClearDepth glColor3d glEnable glEnd glFinish glMatrixMode glNormal3d glPolygonMode glPopMatrix glPushMatrix glRotated glScaled glTranslated glVertex3d glViewport wglCreateContext wglGetCurrentDC wglMakeCurrent IWin32GLU gluCylinder gluDeleteQuadric gluNewQuadric gluPerspective gluQuadricDrawStyle gluQuadricNormals Printer IWin32Printer ClosePrinter DocumentPropertiesA GetPrinterA IWin32PrinterFactory OpenPrinterA OpenPrinterW IWin32PrinterUtility DeviceCapabilitiesA EnumPrintersA Process IWin16ProcessFactory WinExec IWin32Process : IWin32SyncHandle ← IWin32ProcessContext DebugBreak ExitProcess FatalAppExitA FatalExit GetExitCodeProcess GetCurrentProcessId GetProcessVersion GetProcessWorkingSetSize OpenProcessToken SetProcessWorkingSetSize TerminateProcess UnhandledExceptionFilter IWin32ProcessContext GetCommandLineA GetCommandLineW GetCurrentDirectoryA GetCurrentDirectoryW GetStartupInfoA SetConsoleCtrlHandler SetCurrentDirectoryA SetCurrentDirectoryW SetHandleCount SetUnhandledExceptionFilter IWin32ProcessFactory CreateProcessA CreateProcessW OpenProcess Registry IWin16Profile GetPrivateProfileIntA GetPrivateProfileStringA GetPrivateProfileStringW GetProfileIntA GetProfileIntW GetProfileStringA GetProfileStringW WritePrivateProfileStringA WritePrivateProfileStringW WriteProfileStringA WriteProfileStringW IWin16Registry RegCreateKeyExA RegCreateKeyW RegEnumKeyA RegEnumKeyW RegOpenKeyA RegOpenKeyW RegQueryValueA RegQueryValueW RegSetValueA RegSetValueW IWin32Registry RegCloseKey RegCreateKeyA RegCreateKeyExW RegDeleteKeyA RegDeleteKeyW RegDeleteValueA RegDeleteValueW RegEnumKeyExA RegEnumKeyExW RegEnumValueA RegEnumValueW RegFlushKey RegNotifyChangeKeyValue RegOpenKeyExA RegOpenKeyExW RegQueryInfoKeyA RegQueryInfoKeyW RegQueryValueExA RegQueryValueExW RegSetValueExA RegSetValueExW Resource IWin32Resource LoadResource LockResource Security IWin32SecurityACL AddAccessAllowedAce AddAccessDeniedAce AddAce DeleteAce GetAce GetAclInformation IWin32SecurityACLUtility InitializeAcl IsValidAcl IWin32SecurityAccess CopySid EqualSid GetLengthSid IsValidSid LookupAccountNameA LookupAccountSid LookupPrivilegeValueA IWin32SecurityDescriptor GetSecurityDescriptorDacl GetSecurityDescriptorGroup GetSecurityDescriptorOwner GetSecurityDescriptorSacl IsValidSecurityDescriptor SetSecurityDescriptorDacl SetSecurityDescriptorGroup SetSecurityDescriptorOwner SetSecurityDescriptorSacl IWin32SecurityDescriptorFactory InitializeSecurityDescriptor IWin32SecurityToken : IWin32Handle AdjustTokenPrivileges GetTokenInformation IWin32SecurityToken : IWin32Handle OpenProcessToken OpenThreadToken Shell IWin32Drop DragFinish DragQueryFileW DragQueryPoint IWin32Shell SHGetDesktopFolder SHGetFileInfoA ShellExecuteA Synchronization IWin32AtomicUtility InterlockedDecrement InterlockedExchange InterlockedIncrement IWin32CriticalSection DeleteCriticalSection EnterCriticalSection LeaveCriticalSection IWin32CriticalSectionFactory InitializeCriticalSection IWin32Event : IWin32SyncHandle PulseEvent ResetEvent SetEvent IWin32EventFactory CreateEventA IWin32Mutex : IWin32SyncHandle ReleaseMutex IWin32MutexFactory CreateMutexA OpenMutexA IWin32Semaphore : IWin32SyncHandle ReleaseSemaphore IWin32SemaphoreFactory CreateSemaphoreA IWin32SyncHandle : IWin32Handle MsgWaitForMultipleObjects SignalObjectAndWait WaitForMultipleObjects WaitForSingleObject WaitForSingleObjectEx IWin32WaitableTimer : IWin32SyncHandle CancelWaitableTimer SetWaitableTimer IWin32WaitableTimerFactory CreateWaitableTimer OpenWaitableTimer System IWin32WindowsHook CallNextHookEx UnhookWindowsHookEx IWin32WindowsHookFactory SetWindowsHookExA SetWindowsHookExW IWin32WindowsHookUtility CallMsgFilterA CallMsgFilterW Thread IWin32Thread : IWin32SyncHandle ← IWin32ThreadContext, IWin32ThreadMessage DispatchMessageA DispatchMessageW ExitThread GetCurrentThreadId GetExitCodeThread GetThreadLocale GetThreadPriority OpenThreadToken ResumeThread SetThreadPriority SetThreadToken Sleep SuspendThread TerminateThread TlsAlloc TlsFree TlsGetValue TlsSetValue IWin32ThreadContext EnumThreadWindows GetActiveWindow IWin32ThreadFactory CreateThread IWin32ThreadMessage GetMessageA GetMessagePos GetMessageTime GetMessageW GetQueueStatus PostQuitMessage PostThreadMessageA TranslateMessage WaitMessage IWin32ThreadUtility Timer IWin32Timer KillTimer SetTimer Utilities IWin32Beep Beep MessageBeep IWin32StringUtility CharLowerA CharLowerBuffA CharLowerW CharNextA CharNextW CharPrevA CharToOemA CharUpperA CharUpperBuffA CharUpperBuffW CharUpperW CompareStringA CompareStringW FormatMessageA FormatMessageW GetStringTypeA GetStringTypeExA GetStringTypeW IsCharAlphaA IsCharAlphaNumericA IsCharAlphaNumericW IsCharAlphaW IsDBCSLeadByte IsDBCSLeadByteEx LCMapStringA LCMapStringW MultiByteToWideChar OutputDebugStringA OutputDebugStringW ToAscii WideCharToMultiByte lstrcatA lstrcmpA lstrcmpiA lstrcpyA lstrcpyW lstrcpynA lstrlenA lstrlenW wsprintfA wsprintfW wvsprintfA IWin32SystemUtility CountClipboardFormats EmptyClipboard EnumClipboardFormats EnumSystemLocalesA GetACP GetCPInfo GetComputerNameW GetCurrentProcess GetCurrentProcessId GetCurrentThread GetCurrentThreadId GetDateFormatA GetDateFormatW GetDialogBaseUnits GetDoubleClickTime GetLastError GetLocalTime GetLocaleInfoA GetLocaleInfoW GetOEMCP GetSysColor GetSysColorBrush GetSystemDefaultLCID GetSystemDefaultLangID GetSystemDirectoryA GetSystemInfo GetSystemMetrics GetSystemTime GetTickCount GetTimeFormatA GetTimeFormatW GetTimeZoneInformation GetUserDefaultLCID GetUserDefaultLangID GetUserNameA GetUserNameW GetVersion GetVersionExA GetWindowsDirectoryA GetWindowsDirectoryW GlobalMemoryStatus IsValidCodePage IsValidLocale OemToCharA QueryPerformanceCounter QueryPerformanceFrequency RaiseException RegisterWindowMessageA SetErrorMode SetLastError SetLocalTime SystemParametersInfoA IWin32Utility MulDiv Window IWin32Accel CopyAcceleratorTableA TranslateAcceleratorA IWin32AccelFactory LoadAcceleratorsA IWin32Dialog : IWin32Window ← IWin32DialogState ChooseColorA DialogBoxParamA DialogBoxParamW EndDialog MapDialogRect SendDlgItemMessageA IWin32DialogFactory CreateDialogIndirectParamA CreateDialogParamA DialogBoxIndirectParamA IWin32DialogState CheckDlgButton GetDlgCtrlID GetDlgItem GetDlgItemInt GetDlgItemTextA GetNextDlgGroupItem GetNextDlgTabItem IsDlgButtonChecked SetDlgItemInt SetDlgItemTextA IWin32Menu ← IWin32MenuState DeleteMenu DestroyMenu InsertMenuA InsertMenuW IsMenu ModifyMenuA RemoveMenu TrackPopupMenu IWin32MenuFactory CreatePopupMenu IWin32MenuState AppendMenuA AppendMenuW ArrangeIconicWindows BringWindowToTop CheckMenuItem CheckMenuRadioItem CheckRadioButton EnableMenuItem GetMenuItemCount GetMenuItemID GetMenuItemRect GetMenuState GetMenuStringA GetSubMenu HiliteMenuItem SetMenuDefaultItem SetMenuItemBitmaps IWin32Window← IWin32WindowProperties, IWin32WindowState BeginPaint CallWindowProcA CallWindowProcW ChildWindowFromPoint ChildWindowFromPointEx ClientToScreen CloseWindow CreateCaret DefFrameProcA DefMDIChildProcA DefWindowProcA DefWindowProcW DestroyWindow DlgDirListA DlgDirListComboBoxA DlgDirSelectComboBoxExA DlgDirSelectExA DrawAnimatedRects DrawMenuBar EndPaint EnumChildWindows EnumWindows FindWindow FlashWindow MapWindowPoints MessageBoxA MessageBoxW MoveWindow OpenClipboard OpenIcon PeekMessageA PeekMessageW PostMessageA PostMessageW RedrawWindow ScreenToClient ScrollWindow ScrollWindowEx SendMessageA SendMessageW SendNotifyMessageA TranslateMDISysAccel UpdateWindow IWin32WindowFactory CreateWindowExA CreateWindowExW IWin32WindowProperties DragAcceptFiles GetClassLongA GetClassNameA GetClassNameW GetPropA GetPropW RemovePropA RemovePropW SetClassLongA SetPropA SetPropW IWin32WindowState EnableScrollBar EnableWindow GetClientRect GetDC GetDCEx GetLastActivePopup GetMenu GetParent GetScrollInfo GetScrollPos GetScrollRange GetSystemMenu GetTopWindow GetUpdateRect GetUpdateRgn GetWindow GetWindowDC GetWindowLongA GetWindowLongW GetWindowPlacement GetWindowRect GetWindowRgn GetWindowTextA GetWindowTextLengthA GetWindowTextW GetWindowThreadProcessId HideCaret InvalidateRect InvalidateRgn IsWindowEnabled IsChild IsIconic IsWindow IsWindowUnicode IsWindowVisible IsZoomed LockWindowUpdate SetActiveWindow SetClipboardViewer SetFocus SetForegroundWindow SetMenu SetParent SetScrollInfo SetScrollPos SetScrollRange SetWindowLongA SetWindowLongW SetWindowPlacement SetWindowPos SetWindowRgn SetWindowTextA SetWindowTextW ShowCaret ShowOwnedPopups ShowScrollBar ShowWindow ValidateRect ValidateRgn IWin32WindowUtility AdjustWindowRect AdjustWindowRectEx EnumWindows FindWindowA GetActiveWindow GetCapture GetCaretPos GetClassInfoA GetClassInfoExA GetClassInfoW GetDesktopWindow GetFocus GetForegroundWindow InSendMessage IsDialogMessageA RegisterClassA RegisterClassExA RegisterClass

[0092] Although the invention has been described in language specific to structural features and/or methodological steps, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or steps described. Rather, the specific features and steps are disclosed as preferred forms of implementing the claimed invention.

Claims

1. A method of operating an operating system comprising:

defining a translation layer between an application and an operating system, wherein the operating system includes functions that are callable by the application;
receiving calls from the application in the translation layer;
translating the calls with the translation layer to provide translated calls; and
calling operating system functions with the translated calls.

2. The method of claim 1, wherein the defining of the translation layer comprises instantiating a plurality of programming objects representing operating system resources, wherein the programming objects include data and methods, and wherein the methods correspond to operating system functions.

3. The method of claim 1, wherein the defining of the translation layer comprises instantiating a plurality of programming COM objects representing operating system resources, wherein the programming COM objects include data and methods, and wherein the methods correspond to operating system functions, and further wherein object data is accessible through its methods.

4. The method of claim 1, wherein the receiving of the calls comprises receiving calls across at least one process boundary.

5. The method of claim 1, wherein the receiving of the calls comprises receiving calls across at least one machine boundary.

6. The method of claim 1, wherein:

the defining of the translation layer comprises instantiating a plurality of programming objects representing operating system resources, wherein the programming objects include data and methods, and wherein the methods correspond to operating system functions; and
the receiving of the calls comprises receiving calls across at least one process boundary.

7. The method of claim 1, wherein:

the defining of the translation layer comprises instantiating a plurality of programming COM objects representing operating system resources, wherein the programming COM objects include data and methods, and wherein the methods correspond to operating system functions, and further wherein object data is accessible through its methods; and
the receiving of the calls comprises receiving calls across at least one process boundary.

8. The method of claim 1, wherein:

the defining of the translation layer comprises instantiating a plurality of programming objects representing operating system resources, wherein the programming objects include data and methods, and wherein the methods correspond to operating system functions; and
the receiving of the calls comprises receiving calls across at least one machine boundary.

9. The method of claim 1, wherein:

the defining of the translation layer comprises instantiating a plurality of programming COM objects representing operating system resources, wherein the programming COM objects include data and methods, and wherein the methods correspond to operating system functions, and further wherein object data is accessible through its methods; and
the receiving of the calls comprises receiving calls across at least one machine boundary.

10. The method of claim 1, wherein the calls that are received from the application utilize a handle that is not understood by the operating system functions.

11. The method of claim 1, wherein the calls that are received from the application utilize a first handle that is not understood by the operating system functions, and wherein the translating of the calls comprising finding a second handle that corresponds to the first handle and that is understood by the operating system functions.

12. A method of organizing operating system resources comprising:

representing a plurality of operating system resources as programming objects, each programming object representing an operating system resource and having one or more interfaces that define methods that can be used by applications to access and manipulate the resource;
assigning each interface a unique identifier, wherein different versions of a resource are associated with different unique identifiers that can be used by applications to access particular versions of a resource; and
specifying a unique identifier any time an application wants to use a particular resource.

13. The method of claim 12, wherein the unique identifier comprises a globally unique identifier (GUID).

14. The method of claim 12, wherein the unique identifiers represent the syntax of a particular programming object.

15. The method of claim 12, wherein the unique identifiers represent the semantics of a particular programming object.

16. The method of claim 12, wherein the unique identifiers represent the syntax and the semantics of a particular programming object.

17. The method of claim 12, wherein the unique identifiers comprise hexadecimal numbers.

18. The method of claim 12, wherein the specifying of the unique identifier comprises specifying at least one unique identifier across a process boundary.

19. The method of claim 12, wherein the specifying of the unique identifier comprises specifying at least one unique identifier across a machine boundary.

20. The method of claim 12, wherein the specifying of the unique identifier comprises specifying at least one unique identifier across one of a process or machine boundary.

21. The method of claim 12, wherein the specifying of the unique identifier comprises specifying a plurality of unique identifiers, at least some of which being specified across a process boundary, and others of which being specified across a machine boundary.

22. The method of claim 12 further comprising:

determining, based on the unique identifier, whether the operating system has the resource that corresponds to the unique identifier; and
if the operating system does not have the resource that corresponds to the unique identifier, retrieving the resource that corresponds to the unique identifier from a location that can be ascertained by the operating system.

23. The method of claim 22, wherein the location is across at least one machine boundary.

24. The method of claim 22, wherein the location comprises a web site that provides access to code that can be downloaded by the operating system for providing the resource to the application.

25. A computer program embodied on a computer-readable medium that contains instructions which, when executed by a computer, perform the steps comprising:

creating a first handle that is associated with an operating system resource that is represented as a programming object having data and one or more methods;
creating a second handle that corresponds with the first handle;
storing the first handle in the programming object's private state;
passing the second handle to an application instead of the first handle when an application makes calls to create the operating system resource; and
using the second handle to access the operating system resource.

26. The steps of claim 25, wherein the first handle is only valid within its own process, and the second handle is valid outside of the process.

27. The steps of claim 25, wherein the second handle is valid across to process and machine boundaries.

28. The steps of claim 25, wherein the first handle is only valid within its own process, and the second handle is valid across process and machine boundaries.

29. The steps of claim 25, wherein the second handle comprises an interface pointer to an interface of the programming object.

30. The method of claim 25 further comprising determining whether the object is in process, local, or remote before using the second handle to access it.

31. One or more programmed computers having code that contains instructions which, when executed by a computer performs the following steps:

calling an operating system function with a call having a syntax, wherein the call is configured to create or use an operating system resource;
intercepting the call;
changing the syntax of the call to provide a second call with a changed syntax; and
calling a programming object with the second call, wherein the programming object has data and methods and represents the operating system resource.

32. The steps of claim 31 wherein the intercepting comprises interposing a detour function between an application and the operating system, wherein the detour function intercepts every application call to the operating system.

33. The steps of claim 32, wherein the calling of the operating system function comprises doing so across at least one process boundary.

34. The steps of claim 32, wherein the calling of the operating system function comprises doing so across at least one machine boundary.

35. The steps of claim 32, wherein the detour function is configured to enable communication of the call having the appropriate syntax across at least one of a process boundary or machine boundary using DCOM communication techniques.

36. The steps of claim 32 further comprising calling a trampoline that contains code that enables the detour function to remain within its own process.

37. A method of building an application comprising:

establishing a library containing one or more interfaces that are identified by a unique identifier, wherein the interfaces comprise one or more methods that can be used to manipulate programming objects that represent operating system resources; and
linking an application against the unique identifiers.

38. The method of claim 37, wherein the linking of the application takes place against the unique identifiers instead of any dynamic link libraries (DLLs).

39. The method of claim 37, wherein the unique identifier comprises a globally unique identifier (GUID).

40. The method of claim 37, wherein the unique identifier represents the syntax of the interface that it identifies.

41. The method of claim 37, wherein the unique identifier represents the semantics of the interface that it identifies.

42. The method of claim 37, wherein the unique identifier represents the syntax and the semantics of the interface that it identifies.

43. A computer readable medium having computer-readable instructions thereon, which when executed by a computer perform the following steps:

representing a plurality of operating system resources as objects having one or more interfaces through which one or more resources can be accessed;
assigning each interface a globally unique identifier (GUID) that can be used by an application to access a particular resource, wherein the GUIDs are assigned so that they distinguish between different versions of the same object and represent the syntax and semantics of an object; and
specifying the unique identifier across at least one process or machine boundary any time an application wants to use a particular object.

44. The steps of claim 43 further comprising determining, based on the unique identifier, whether the operating system has the object that corresponds to the unique identifier and if not, ascertaining the location of the object and retrieving the object from the location.

45. One or more computer-readable media comprising:

a library of operating system functions;
a plurality of programming objects representing respective operating system resources, each programming object comprising data and associated methods, wherein the data is accessible only through the methods;
wherein the methods of the programming objects are callable to access the resources represented by the objects; and
wherein the methods of the programming objects are responsive to call the functions of the function library.

46. The computer readable media of claim 45 further comprising a plurality of pairs of first and second handles, wherein the handle pairs correspond to one another, and wherein the first handles of each pair are usable by one or more applications to call the programming objects, and wherein the second handles of each pair are usable by the objects to call the functions of the function library.

47. The computer readable media of claim 45, wherein the programming objects are distributed across one or more process or machine boundaries.

48. The computer readable media of claim 45, wherein the media comprise a computer network comprising a plurality of computers connected for data communication.

49. A method of utilizing an existing function library of an operating system comprising:

defining a plurality of programming objects representing respective operating system resources, each programming object comprising data and associated methods, wherein the data is accessible only through the methods;
calling the methods of the programming objects to access the resources represented by the objects; and
calling the functions of the function library from the programming object methods.

50. The method of claim 49, wherein said calling steps are performed using two different handles that are associated with one another.

51. The method of claim 49, wherein the calling of the methods comprises first calling a detour using a first call syntax, using the detour to change the first call syntax into a second call syntax that is different from the first call syntax, and second calling the methods using the second call syntax.

52. The method of claim 49, wherein the calling of the methods comprises doing so across process and machine boundaries.

53. The method of claim 49, wherein the calling of the methods comprises first calling a detour from an application and second calling a trampoline from the detour, wherein the detour contains code that enables programming objects within a current process to be called, and wherein the trampoline is callable by the detour and contains code to enable the detour to stay within its own process.

54. A method of utilizing an object-oriented operating system function library from a legacy application program that is configured to utilize non-object-oriented functions of an operating system function library, comprising:

defining a plurality of detour functions corresponding to the functions of the operating system function library;
calling a detour function from a legacy application;
calling a programming object from the detour function, wherein the programming object has data and methods; and
calling a function in the function library using one of the methods of the programming object.

55. The method of claim 54, wherein the calling of the detour function comprises using a first handle that corresponds to an operating system resource, and wherein the calling of the function comprises using a second handle that corresponds to and is different from the first handle.

56. A computer-readable medium comprising:

a plurality of programming objects that represent operating system resources, wherein the objects contain data and methods, and wherein the data is accessible through the methods;
wherein the programming objects are configured to receive calls that call the methods; and
wherein at least some of the methods are configured to call functions of an operating system to manipulate operating system resources.

57. The computer readable media of claim 56 further comprising a detour that is configured to receive a call from an application that is intended for an operating system function, and responsive thereto call a method on one of the programming objects.

Patent History
Publication number: 20030177285
Type: Application
Filed: Mar 10, 2003
Publication Date: Sep 18, 2003
Patent Grant number: 7334235
Inventors: Galen C. Hunt (Bellevue, WA), Gerald Cermak (Bothell, WA), Robert J. Stets (Rochester, NY)
Application Number: 10385381
Classifications
Current U.S. Class: 709/328
International Classification: G06F009/46; G06F009/00;